Organic light-emitting diodes (OLEDs) have been widely used, particularly in display applications. OLEDs are easily degraded without stringent encapsulation owing to their susceptibility to water vapor and oxygen. Therefore, establishing an effective protection method for these devices is essential. In this study, we demonstrate the device protection performance and improvement in color purity by introducing CaF2/ZnS multilayered films on a top-emitting inverted-type OLED (iOLED), which was originally intended to act as a distributed Bragg reflector (DBR). To test the protection performance of each dielectric layer, conventional bottom-emitting OLEDs (cOLEDs) with and without single layers of CaF2 and ZnS were investigated for comparison. All OLEDs were stored in an atmosphere without stringent encapsulation, such as a cover glass. The luminescence area of cOLEDs without the dielectric film decreased by more than 90% after 3 days of fabrication. In contrast, the dark-spot formation was moderated after the same period for the dielectric single-layer deposited cOLEDs. Notably, the iOLED with DBR completely preserved the emitting area even after 2 months of fabrication. This suggests that DBR acted as a protective film for the organic layer, whereas the inverted structure also contributed to reducing the degradation of air- and moisture-sensitive materials.
Organic light-emitting diodes (OLEDs) have been widely used, particularly in display applications. OLEDs are easily degraded without stringent encapsulation owing to their susceptibility to water vapor and oxygen. Therefore, establishing an effective protection method for these devices is essential. In this study, we demonstrate the device protection performance and improvement in color purity by introducing CaF2/ZnS multilayered films on a top-emitting inverted-type OLED (iOLED), which was originally intended to act as a distributed Bragg reflector (DBR). To test the protection performance of each dielectric layer, conventional bottom-emitting OLEDs (cOLEDs) with and without single layers of CaF2 and ZnS were investigated for comparison. All OLEDs were stored in an atmosphere without stringent encapsulation, such as a cover glass. The luminescence area of cOLEDs without the dielectric film decreased by more than 90% after 3 days of fabrication. In contrast, the dark-spot formation was moderated after the same period for the dielectric single-layer deposited cOLEDs. Notably, the iOLED with DBR completely preserved the emitting area even after 2 months of fabrication. This suggests that DBR acted as a protective film for the organic layer, whereas the inverted structure also contributed to reducing the degradation of air- and moisture-sensitive materials.
In recent years, organic
light-emitting diode (OLED) displays have
attracted considerable interest because of their advantages, such
as fast response time, light weight, low cost, and flexibility.[1−3] In addition to these advantages, OLEDs are expected to emerge as
mainstream displays in the future because they can be fabricated on
flexible substrates.[4−8] However, OLEDs have the disadvantage of the rapid degradation of
the materials and reduction of the light-emitting area. The materials
(LiF and Al) used in the device structure for electrodes and the electron
injection layer are easily degraded by atmospheric oxygen and water
vapor. Consequently, a non-light-emitting area called a dark spot
is formed without stringent encapsulation.[9] Therefore, it is necessary to develop high-performance thin-film
encapsulation technology to realize an OLED display that can be driven
in the atmosphere for a long time. Currently, the most commonly used
method is to fabricate an OLED and then rigorously encapsulate it
with a cover glass using an ultraviolet (UV)-epoxy resin. The UV-epoxy
resin cannot completely prevent the ingress of water vapor in the
atmosphere; therefore, a strong desiccant must be sealed with an OLED.
Although this method of encapsulation applies to bottom-emission OLEDs,
an opaque desiccant may inhibit emission extraction in top-emission
OLEDs. Furthermore, when using a flexible substrate for OLEDs, encapsulation
films and desiccants require flexibility. Therefore, it is necessary
to develop methods for encapsulating and preventing degradation that
can be applied to flexible substrates.[10]Improvements in the device lifetimes of OLEDs using inorganic
materials,
polymers, and inorganic/organic hybrid materials have been reported.[10−15] Kim et al. used polyacrylate as a protective layer in a spin-coating
method and found that the lifetime of the device reached 12.6 h.[12] Ogawa et al. confirmed that SiN/SiON fabricated via catalytic chemical vapor deposition (Cat-CVD)
protects OLEDs at 60 °C and 90% RH for 1000 h,[14] and Ghosh et al. confirmed that an Al2O3 layer fabricated using atomic layer deposition (ALD) protects
OLEDs at 85 °C and 85% RH for 1300 h.[15] As in these studies, Cat-CVD and ALD are widely used to produce
high-quality protective films.[12−15] However, high-temperature deposition with substrate
heating is usually necessary while using these deposition methods,
and it may cause damage to the organic layers during the deposition
process. In contrast, thermal evaporation can be suitable because
it is a relatively gentle deposition method with comparatively low
kinetic energy, without introducing plasma in the chamber or corrosive
chemical reaction. If these sealing films can be formed using the
simple vacuum deposition method, which is facile and is also used
to form organic layers, it may be possible to realize long-life device
structures with a simpler manufacturing process, considering the affinity
toward the existing deposition processes.Another issue that
needs to be considered in OLED display applications
is color purity. The triangles formed by the red, blue, and green
vertices in the color coordinates indicate the color area that can
be represented in a display. The range of colors that can be represented
needs to be expanded to display an image closer to what the eye can
see. However, improvements are required because the organic light-emitting
materials used in conventional OLEDs (cOLEDs) generally have a wide
spectral range, and the mixing of various light wavelength components
results in low color purity and a narrow range of colors that can
be expressed. Currently, two approaches are being explored for improving
the color purity: using properly designed light-emitting materials
and using resonators. For example, the modification of the molecular
design from Ir(ppy)3 to Ir(mtfppy)3, a phosphorescent
luminescent material, resulted in a narrower emission spectrum and
improved color purity in the green region.[16] In another method, an optical resonator (cavity) composed of two
mirrors was incorporated into the OLED device. By adjusting the cavity
length (optical film thickness) between the mirrors, a certain wavelength
of light can be selectively extracted, resulting in a narrow electroluminescence
(EL) spectrum and improved color purity.[17−22]In our previous study, the color purity of green OLEDs was
improved
by introducing a distributed Bragg reflector (DBR) fabricated via
vacuum deposition on top of an OLED device with an inverted structure.[23] The previous report focused only on improving
the color purity, but herein, we consider that this multilayer film
could also work as a protective layer for OLEDs. Selection of the
dielectric materials such as calcium fluoride (CaF2) for
low-index and zinc sulfide (ZnS) for high-index is suitable for depositing
the multilayered films on the OLED device because they can deposit
by the thermal vacuum evaporation method at room temperature with
enough optical quality.[24] The room temperature
deposition also allows for application to the encapsulation of OLEDs
on flexible substrates with low heat tolerance. In this study, we
performed three experiments: (1) the deposition of a single layer
of a dielectric material (CaF2 and ZnS) on a cOLED to investigate
the protection performance of each layer; (2) the investigation of
the impact of using a reversing structure on the dark-spot formation,
which makes the layer prone to degradation and less exposed to the
atmosphere; and (3) the durability test of the OLED device with both
dielectric multilayers (DBR) and a reversed structure. The results
showed that the inverted OLED structure with DBR prevented dark-spot
formation for more than 2 months after fabrication, even without further
encapsulation. This is probably because of the inverted structure
and film formation of the DBR covering the OLED. In terms of color
purity, the full width at half maximum (fwhm) of the EL spectrum of
the DBR-based OLED could be significantly narrowed from 70 to 19 nm
owing to the microcavity effect.
Materials
and Methods
The cOLEDs were fabricated on the commercially
available glass
substrates coated with an indium tin oxide (ITO) layer that was ultrasonically
cleaned with acetone and isopropyl alcohol, followed by UV ozone cleaning.
Inverted structure OLEDs (iOLEDs) were fabricated on glass substrates
that were ultrasonically cleaned with detergent, purified water, and
ethanol, followed by UV ozone cleaning. The detailed sample structures
and the geometry, size, and order of each layer of cOLED, iOLED, and
μ-cavity iOLED are shown in Figure .
Figure 1
Fabricated device structures of green OLEDs:
(a) bottom-emission
cOLED with and without a single dielectric layer of CaF2 and ZnS. (b) Top-emission iOLED. (c) μ-cavity iOLED with different
pairs of CaF2/ZnS DBR. Geometry, size, and order of each
deposited layer of (d) cOLED and (e) iOLED and μ-cavity iOLED.
Fabricated device structures of green OLEDs:
(a) bottom-emission
cOLED with and without a single dielectric layer of CaF2 and ZnS. (b) Top-emission iOLED. (c) μ-cavity iOLED with different
pairs of CaF2/ZnS DBR. Geometry, size, and order of each
deposited layer of (d) cOLED and (e) iOLED and μ-cavity iOLED.All the organic and dielectric films as well as
the metal electrodes
fabricated in this study were prepared using the thermal vacuum evaporation
method at room temperature without substrate heating. The entire process
was done in a high vacuum chamber (<4 × 10–4 Pa). During the deposition, the film thickness and deposition rate
were controlled by monitoring the quartz crystal microbalance. Ag
and Al wires (4 N grade) were employed as the anode and cathode materials,
respectively. As the light-emitting layer, tris(8-hydroxyquinolinato)aluminum
(Alq3) was used for the green-emitting OLED. LiF, MoO3, and N,N′-di-1-naphthyl-N,N′-diphenylbenzidine (α-NPD)
were used for the electron injection, hole injection, and hole transport
layers, respectively. Calcium fluoride powder (CaF2, 99.9%
purity) and zinc sulfide powder (ZnS, 99.999% purity) were used for
the evaporation source of dielectric layers. The film thickness of
the DBR fabricated in this study was determined to satisfy Bragg’s
conditions with the target wavelength of 535 nm for the center of
stopband.The device characteristics were measured using a source
measure
unit (Keithley 2401) and a chromameter (Konica Minolta, CS-200). A
multi-channel spectrometer (Ocean Optics, USB4000) was used for recording
the EL spectra. The fabricated OLEDs were stored in a desiccator without
being strictly sealed with a cover glass and were not energized to
observe the changes over time.
Results and Discussion
Protection Performance of a Single-Layer Dielectric
Film
First, we examined the protection performance of the
single-layer dielectric film against the formation of dark spots in
cOLEDs without stringent encapsulation using a cover glass. Single
layers of CaF2 (200 nm) and ZnS (200 and 500 nm) were deposited
on cOLED devices. The cOLED device structure was ITO/α-NPD (60
nm)/Alq3 (60 nm)/LiF (1 nm)/Al (150 nm). Figure a shows images of the changes
in the emitting area of each OLED device, and the emissive area of
each device is plotted against the time of fabrication.
Figure 2
(a) Photographs
of the emission pattern of cOLED with and without
single-layer dielectric films taken immediately and 3 days after the
fabrication. (b) Time-dependent changes in the emissive area as a
function of days of cOLED with and without single-layer dielectric
films.
(a) Photographs
of the emission pattern of cOLED with and without
single-layer dielectric films taken immediately and 3 days after the
fabrication. (b) Time-dependent changes in the emissive area as a
function of days of cOLED with and without single-layer dielectric
films.Because the cOLEDs were not stringently
encapsulated, the dark
spots were spread over the entire area of the device, resulting in
a 90% decrease in the emissive area without dielectric coating 3 days
after the fabrication. This is probably because of the degradation
of Al and LiF located near the top side of the device. In contrast,
by depositing a 200 nm CaF2 or ZnS single layer protection
film on top of the cOLED, the dark-spot formation was moderated because
the dielectric material covered the device, which inhibited contact
with oxygen and water vapor that caused degradation. There was also
a difference in the protection performance between the materials;
the emitting area using the CaF2 film was reduced to 38%
after 3 days, whereas it was retained at 80% using ZnS, indicating
a better protection performance of ZnS compared to that of CaF2. This is consistent with a previous report that stated that
ZnS is more resistant to heat and the environment.[10] Furthermore, by increasing the film thickness, the ZnS
500 nm device retained 90% of its emissive area even after 3 days
of fabrication, indicating that the protective performance was further
improved.As shown in Figure , the emissive area is the 2 × 5 mm2 section, where
both the electrodes overlap each other. When a dielectric protective
layer is deposited, at least 2 to 3 mm of the area around the emissive
area is surely covered by the dielectric layer. In the case of protection
performance with CaF2, as shown in Figure a, the area around the emissive area is noticeably
darkened, but at least when ZnS encapsulation is used, the generation
of non-emissive areas from the surroundings is significantly suppressed.
Therefore, the dielectric film coating prevents the deterioration
of the device due to water vapor and oxygen penetration from the side
of the films.
cOLED vs iOLED
Next, we investigated
the effectiveness of adopting an inverted structure to suppress dark-spot
formation, where it is more difficult to expose the moisture-sensitive
layers (Al and LiF) to the atmosphere, by swapping the positions of
the anode and cathode. The effect of the inverted structure on the
device performance was reported by Fukagawa et al.[25] They succeeded in suppressing the degradation of OLEDs
by employing an inverted structure using the bottom ITO as the cathode.
By developing electron injection materials, a device performance comparable
to that of a cOLED with a longer lifetime was achieved for the iOLED
even when using the bottom ITO cathode, and no change in the EL spectrum
was observed.In the present study, we did not change the material
of the Al cathode but simply swapped the position of the bottom and
top surfaces to reduce the degradation of the Al and LiF layers by
avoiding exposure to the atmosphere. In addition, we employed a Ag
thin-film anode instead of ITO for the top electrode, thereby expecting
an improvement in conductivity and color purity owing to the microcavity
effect between the reflective electrodes. The device structure consisted
of Al (150 nm)/LiF (1 nm)/Alq3 (60 nm)/α-NPD (60
nm)/MoO3 (2 nm)/Ag (14 nm). Figure shows the protection performance against
the dark-spot formation.
Figure 3
(a) Photographs of the emission pattern of cOLED
and iOLED taken
immediately and 3 days after the fabrication. (b) Time-dependent changes
in the emissive area as a function of days of cOLED and iOLED.
(a) Photographs of the emission pattern of cOLED
and iOLED taken
immediately and 3 days after the fabrication. (b) Time-dependent changes
in the emissive area as a function of days of cOLED and iOLED.No dark spots were observed for the iOLED even
after 3 days of
fabrication. This indicates that the Al and LiF layers, which are
prone to degradation, have moved to the bottom of the device in the
inverted structure, providing more effective protection compared to
the dielectric layer coating. The device performance of the iOLED
was affected by swapping the positions of the electrodes and employing
the Ag anode, as shown in Figure a–c and Table .
Figure 4
(a) J–V (current
density–voltage)
and (b) L–V (luminance–voltage)
characteristics and (c) current efficiency of cOLED (black open circle),
iOLED (blue open circle), and μ-cavity iOLEDs with 1 (green
filled circle), 3 (red filled circle), and 5 (yellow filled circle)
pairs of DBR.
Table 1
Device Performance
of cOLED, iOLED,
and μ-Cavity iOLEDs with 1, 3, and 5 Pairs of DBR
cOLED
iOLED
1 pair
3 pairs
5 pairs
current
efficiency at 50 mA/cm2 (cd/A)
2.79
2.63
4.46
3.23
0.34
external quantum efficiency at 20 mA/cm2
1.54%
0.83%
1.36%
0.88%
0.08%
(a) J–V (current
density–voltage)
and (b) L–V (luminance–voltage)
characteristics and (c) current efficiency of cOLED (black open circle),
iOLED (blue open circle), and μ-cavity iOLEDs with 1 (green
filled circle), 3 (red filled circle), and 5 (yellow filled circle)
pairs of DBR.From the current density–voltage–luminance
(J–V–L) characteristics
shown in Figure a,b,
the threshold voltage of the iOLED was increased to 4.0 V compared
to 2.8 V of the cOLED. This increase probably originated from the
difference in the work function of ITO and Ag by inserting the MoO3 layer, which was already confirmed in the study of green
OLED with Ag electrodes,[26] whereas the
interface roughness of the electrode can also be a factor affecting
the work function because the film position (and the order of deposition)
of the anode is different for the cOLED and iOLED. However, the current
efficiency of the iOLED is comparable to that of the cOLED, as shown
in Figure c. This
indicated that the recombination process itself was not affected by
the introduction of the inverted structure.
μ-Cavity
iOLED
In this section,
we describe the fabrication of a μ-cavity iOLED by depositing
a dielectric DBR on the top surface of iOLED to improve the color
purity as well as the protection performance of the device. The thickness
of the CaF2/ZnS DBR was designed for the emission peak
at 520 nm, and the number of pairs was prepared as 1, 3, and 5. The
thicknesses of the CaF2/ZnS films were set to 94 nm/56
nm, respectively. Figure a–c shows the device characteristics of the fabricated
OLEDs. There was a slight increase in threshold voltages among the
μ-cavity iOLEDs compared to the cOLED, whose values were close
to those of the iOLED without DBR from the J–V curves (Figure a). This is reasonable because the deposition position of
the DBR is outside the device and may not affect the electrical properties.
In the (luminance–voltage) L–V characteristics, the luminance of the three- and five-pair
DBR μ-cavity OLEDs showed a slightly lower performance, which
probably reflects the spectral narrowing and the matching of the maximum
luminosity function and EL peak. When we compared the current efficiency,
there was negligible change upon changing the structure from cOLED
to iOLED, as mentioned in the previous section, but the OLED device
with one or three pairs of DBR outperformed the cOLED. This may be
because of the contribution of the microcavity effect induced by the
outer DBR mirror and bottom Al electrode. Figure shows the EL spectra and color coordinates.
Figure 5
(a) EL
spectra of cOLED (black), iOLED (blue), and μ-cavity
iOLEDs with 1 (green), 3 (red), and 5 (yellow) pairs of DBR. EL spectra
were normalized at the EL peak intensity. (b) Color coordinates in
CIE1931 color space of cOLED, iOLED, and μ-cavity iOLEDs with
1, 3, and 5 pairs of DBR, and the inset shows the photographs of the
emission pattern of cOLED and μ-cavity iOLEDs with three pairs
of DBR.
(a) EL
spectra of cOLED (black), iOLED (blue), and μ-cavity
iOLEDs with 1 (green), 3 (red), and 5 (yellow) pairs of DBR. EL spectra
were normalized at the EL peak intensity. (b) Color coordinates in
CIE1931 color space of cOLED, iOLED, and μ-cavity iOLEDs with
1, 3, and 5 pairs of DBR, and the inset shows the photographs of the
emission pattern of cOLED and μ-cavity iOLEDs with three pairs
of DBR.Three and five pairs of DBR showed
significantly narrowed spectra
at 520 nm, which is consistent with the designed cavity length and
exhibits high wavelength selectivity. The fwhm of the peaks of EL
spectra is summarized in Table .
Table 2
fwhm of the Peaks of EL Spectra and
Color Coordinates of CIE for cOLED, iOLED, and μ-Cavity iOLEDs
with 1, 3, and 5 Pairs DBR
cOLED
iOLED
1 pair
3 pairs
5 pairs
fwhm (nm)
93
69
86
19
12
CIE (X, Y)
(0.31, 0.54)
(0.32, 0.61)
(0.44, 0.53)
(0.20, 0.72)
(0.24, 0.69)
The spectral width of the main peak
was significantly narrowed
to 12 nm for the iOLED with five pairs of DBR compared to 93 nm for
the cOLED. The microcavity effect is effective in narrowing the peaks
of the EL spectra as the number of DBR pairs increases. The EL spectral
peaks of the iOLED without DBR were also narrower than those of the
cOLED, which is probably because of the microcavity effect between
the Ag and Al electrodes. The one pair case showed an exceptional
result of broadened and red-shifted EL spectra, although the current
efficiency was better than that of other iOLEDs. Comparing the color
purity of the EL spectra in Figure b, the μ-cavity iOLED with three pairs of DBR
showed the highest color purity for green emission. According to the
fwhm data of the EL spectra, the iOLED with five pairs of DBR is expected
to possess the best color purity because the peak wavelength is set
to 520 nm. However, a sub-peak was observed at 650 nm in the EL spectrum
of the iOLED with five pairs of DBR. This was due to the transmission
of emission slightly outside the stopband of the DBR because the spectral
edge of the stopband becomes steeper with an increase in the number
of pairs.Figure shows the
changes in the device characteristics of the three-pair DBR μ-cavity
iOLED over time. In the J–V–L characteristics, the current and luminance
values gradually decrease with time in the high voltage range, but
they are maintained at a certain degree in the low voltage region.
Figure 6
Time-dependent
changes in (a) J–V (current
density–voltage) and (b) L–V (luminance–voltage) characteristics
and (c) current efficiency of μ-cavity iOLEDs with 3 pairs of
DBR.
Time-dependent
changes in (a) J–V (current
density–voltage) and (b) L–V (luminance–voltage) characteristics
and (c) current efficiency of μ-cavity iOLEDs with 3 pairs of
DBR.As shown in Figure c, the current efficiency was maintained
for up to 3 weeks after
fabrication, indicating that the degradation of the OLED was suppressed.
However, gradual degradation was observed after 2 months. One possible
reason for this decrease in the device performance is the degradation
of the Ag electrode. Because the Ag electrode for the contact was
not protected by the dielectric layer in our device design, exposure
to air caused agglomeration, resulting in a discontinuity in the film.
This can increase the resistance of the film, resulting in a decrease
in the luminous efficiency.Figure shows the
images of the emitting area of the μ-cavity iOLED with three
pairs of DBR, showing that a negligible number of dark spots were
formed even 2 months after the fabrication, without stringent encapsulation,
such as using a cover glass.
Figure 7
Photographs of the emission pattern of μ-cavity
iOLEDs with
3 pairs of DBR taken after deposition and 3 days and 2 months after
the device fabrication.
Photographs of the emission pattern of μ-cavity
iOLEDs with
3 pairs of DBR taken after deposition and 3 days and 2 months after
the device fabrication.This protection against
dark-spot formation was achieved by coating
the dielectric multilayer film on top of the device and employing
an inverted-type device structure. In this μ-cavity iOLED structure,
the CaF2/ZnS dielectric multilayer film acts as a DBR and
a protective layer from the atmosphere. The materials used in this
study can be deposited via thermal vacuum evaporation, which is commonly
used for fabricating organic layers and metal electrodes in OLEDs.
Therefore, it is easier to deposit multifunctional layered films for
device protection and improving performance. Further extension of
the device lifetime is expected by combining it with the existing
encapsulation method.
Conclusions
Different
pairs of CaF2/ZnS multilayered films were
deposited on the top-emitting inverted-type OLED, and their optical
properties were examined as DBR for improving the color purity. In
addition, the role of the protective dielectric coating in inhibiting
dark-spot formation in the emissive area was studied. First, we investigated
the device protection performance of a single layer of the dielectric
material (CaF2 and ZnS) deposited on a cOLED and confirmed
its ability to inhibit dark-spot formation to a certain degree. Second,
we investigated the impact of using the inverted structure of the
OLED, where the positions of the anode and cathode were swapped, on
the dark-spot formation. The iOLED showed a significant inhibition
of dark-spot formation by reducing the exposure of the oxygen-sensitive
layer to moisture and/or atmosphere. Finally, we fabricated μ-cavity
iOLEDs with different pairs of DBR, which possess both confirmed protection
characteristics. The results showed that the iOLED with three pairs
of DBR prevented dark-spot formation for more than2 months after fabrication,
even without a stringent encapsulation process. In terms of color
purity, the full width at half maximum of the peaks of the EL spectrum
of the DBR-based OLED could be significantly narrowed from 70 to 19
nm owing to the microcavity effect. Depositing multilayered dielectric
films on the top-emitting iOLED using a simple thermal vacuum evaporation
method, which is a familiar method for fabricating OLED devices, will
be a convenient way to provide the multi-functionality of color purity
and device lifetime enhancements.